In Image Mode, the SAR obtained strips of high-resolution imagery, 100 km in width, to the right of the satellite track. AMI could only operate in one mode at a time. Therefore, operating in Image Mode prevented the operation of the other AMI modes. Power considerations limited the SAR operating time to a maximum of about 12 minutes per orbit. The data rate of 105 Mbit/s was too high to allow on-board storage, and so images were only acquired if the satellite was within range of receiving ground stations.
In Wave Mode, the SAR measured the ocean radar reflectivity over 5 km by 5 km imagettes at intervals of 200 km along track. The imagettes were used to provide ocean spectra, which gave information about the wavelength and direction of ocean wave systems.
More information on the Image and Wave modes is provided below:
ERS-2 Attitude Control Modes and Orbit Dead-band control
The piloting of the ERS-2 satellite was initially performed using three gyroscopes (out of six available), with the number of functioning gyroscopes reducing during the mission lifetime due to failures. SAR products have been used to monitor the Doppler centroid frequency of imagery acquired by ERS-2 as shown in the figure below.
It can be seen that there were several occasions during the early years of the ERS-2 mission, there was a larger than expected change in Doppler centroid frequency. This was due, primarily, to imagery acquired during orbit manoeuvres (imagery acquired during a manoeuvre can still be processed but it may be unsuitable for interferometry). The negative Doppler Centroid frequencies are from ascending pass imagery while positive values are from descending pass imagery. Each product header contains the Doppler information for the product.
Due to the continuing gyroscope failures, a mono-gyro (1GP) piloting mode began operation on 11 February 2000. A consequence of this change was a loss of attitude stability but the impact on SAR product quality was small. For the 1GP mode the Doppler Centroid frequency increased for localised areas and time intervals but values were mostly limited to the first ambiguity number (-1600Hz to 1600Hz). During this IGP period, new on-board software to pilot the satellite without using any gyroscopes by exploiting the capabilities of the Digital Earth Sensor and Digital Sun Sensor was developed and tested. A first coarse version of this software, the Extra Backup mode (EBM) was uploaded to ERS-2 in December 2000 as a first step for gyro-less operations and was operationally used since mid-January 2001 due to further problems with the remaining gyroscopes. Significant degradation of the ERS-2 attitude (particularly for yaw) occurred during the EBM. Very large Doppler Centroid frequencies are difficult to estimate in SAR processor – particularly for values outside ±4500Hz with the ESA SAR processors.
Improvements in attitude control took place in June 2001 with the introduction of the Zero Gyro Mode (ZGM) and again in March 2002 with ZGM-Yaw Control Monitoring (ZGM-YCM) – 90% of the Doppler Centroid frequencies remained within ±4500Hz. A further refinement involved deriving yaw values in real-time from wave-mode data to automatically correct the instrument attitude during the next orbit pass if necessary. Some changes to the strategy were required after the failure of the on-board tape recorder in June 2003, since wave mode acquisitions all around the orbit could not be used anymore. Using a network of receiving ground stations, it was possible to continue receiving wave-mode data although coverage was reduced to the northern hemisphere. This strategy was called Regional YCM (R-YCM).
Subset of ERS-2 Doppler Centroid Frequency over the mission lifetime. The start of Zero Gyro Mode operation in 2001 is clearly visible.
Doppler Centroid Information is available in the links below for the different ERS-2 modes:
A full description of gyro operational status and impact is provided in a paper here. This paper also gives more details of the changes in the interferometric dead-band indicated below due to the impact of gyroscope failures.
Instrument Calibration - Replica pulse power correction factor
A small number of products generated within the ESA ground segment (< 1%) are processed with a nominal replica pulse rather than a replica generated at the time of imaging (an extracted replica). This occurs when the extracted replica pulse is corrupt. Products generated with a nominal replica have significantly higher pixel values than products generated using an extracted replica. These products can be corrected. For ERS-2 SAR products, the correction depends on the extracted replica pulse power at the time of data acquisition. This can be estimated from the quarterly averaged values given in the table below. For 2003 Q1, two corrections are given, one before 26th February and the other after 28th February due to the gain change between these dates (increase of the Image up-converter level by 2.5dB on 26 February 2003 and increase of the image receiver attenuation gain by 1dB on 28 February 2003). Replica pulse powers between 4th September 10:04:14 and 14th October 2004 14:37:11 UTC have been increased by 4dB to correct for the erroneous calibration attenuation gain setting during this period. The correction factor is applied such that the image intensity values need to be reduced. Note that as the correction given in the table has been derived from replica pulses from the start of imaging sequences, no account is made for any variations in replica power between the start and end of imaging sequences.
For more details please see the technical note ERS calibration Derivation of the backscattering coefficient sigma nought in ESA ERS SAR PRI products.
Quarterly averaged nominal replica pulse correction for the ERS-2 SAR (dB)
ERS-2 EBM Mode (from 01/2001 to 06/2001)
Hight rate Doppler available at ESRIN for EBM:
ERS-2 Zero-Gyro Mode (from 06/2001)
Doppler frequency for cycle XX
These plots show the Doppler Centroid frequency values for ERS-2 SAR data acquired during this cycle, separately for ascending and for descending passes. The estimated accuracy of these values is around +/-500 Hz. These plots are intended to provide a geographical overview of the Doppler Centroid frequency values for data acquired during the cycle. For a detailed knowledge of the Doppler Centroid frequency for data acquired over a specific orbit and frame, please goto the interactive Doppler query page or download and consult the Doppler ASCII file for this cycle.
Doppler frequency difference cycle XX & cycle XY
These plots show the estimated absolute Doppler Centroid difference between data acquired in this cycle and in the previous one over the same track and frame. The range of ABS Doppler frequency differences shown here is limited to the range 0-1000 Hz in order to highlight suitable InSAR pairs. Values outside this range appear in red. It should be recalled that the accuracy of the Doppler frequency values used to compute this difference is roughly +/-500 Hz. For a detailed knowledge of the Doppler Centroid frequency difference between a set of products, please download the corresponding Doppler ASCII files.
Doppler frequency values per orbit and frame
The file below provides the estimated Doppler Centroid frequency values for all the orbits and standard frames during this cycle. The accuracy of these values is around +/-500 Hz.The file contains the following parameters, in the order given below: Cycle number, ASC/DESC, Orbit number, UTC time, Frame number, Track number, Centre Latitude, Centre Longitude, Doppler Centroid frequency.Download ASCII file (less then 6Mb)